THE ECOLOGY OF SPELLING: A LENS MODEL ANALYSIS OF SPELLING ERRORS AND STUDENT JUDGMENTS OF SPELLING DIFFICULTY

Abstract

Students who were identified by their teachers as poor spellers were asked to judge the difficulty they would have in spelling each of 100 words that were representative of the sorts of words they tended to misspell in their various subject areas (the spelling ecology). After making difficulty judgments, the students were then asked to spell each word on the list. Spelling errors were scored as either phonetic or nonphonetic. The researchers rated each of the 100 words on ten characteristics: number of letters, syllables, letters per syllable, double and silent letters; schwa, ambiguous, and unusual sounds; and two measures of familiarity to the student. This task was replicated after a four‐week period to check for spelling and judgment consistency. Spelling errors and judgments of spelling difficulty were analyzed using the double system Lens Model using the ten word characteristics as “cues” in the analysis. Results showed only moderate agreement between difficulty judgments and spelling errors, and fairly consistent differences between those word characteristics that were predictive of perceived spelling difficulty and those predictive of phonetic and nonphonetic errors. Several different patterns of cue weights were noted for spelling errors whereas spelling difficulty judgments were primarily based upon word familiarity. Implications are drawn for the further investigation of spelling errors and of how students decide what constitutes a “hard” word to spell and for the potential improvement of the spelling judgment process using cognitive feedback from the Lens Model.

The characterization of spelling as a cognitive activity has received increasing attention in the educational research literature. For example, two recent issues of Reading Psychology (Numbers 2 and 3, 1989) were entirely devoted to research on spelling. Much of this work has focused on spelling strategies (e.g., Anderson, 1985, Kreiner & Gough, 1990, and Olson, Logan, & Lindsey, 1989), spelling problems and types of errors (e.g., see Frith, 1980, and Weber & Henderson, 1989), and the relationship between spelling and reading (e.g., Templeton, 1989, and Zutell & Rasinski, 1989). However, there has been very little research on the issues of the criteria individuals use in deciding that a particular word is difficult or easy to spell, and the accuracy and effects of such a decision. We refer here to the more metacognitive aspects of the task ecology of spelling, and the effects of which may be shown in decisions to avoid or at least minimize the cognitive efforts expended in spelling a perceived “hard” word.

The issue of judging spelling difficulty forms the focus of the present study. Here we attempt to document those features of a word that make students think it would be hard or easy to spell. We then relate this judgment process to errors made when students attempt to spell words. Finally, we compare the relative importance of various features of words as predictors of both difficulty judgments and actual spelling errors.     

Phonological spelling

A case study is presented of phonological spelling, an acquired spelling disorder in which the primary symptom is the occurrence of phonologically plausible errors (e.g. "flood" → flud). Not all of the patient's spelling errors are as phonologically "perfect" as this example; but it is arguable that the errors primarily derive from a routine which segments a phonological code and assigns orthographic representations to these individual segments. This account of errors in phonological spelling is contrasted with an interpretation of oral reading errors in surface dyslexia. We conclude that errors in the two disorders do not reveal a precise parallel, and that the contrast is partly attributable to the differential role of comprehension in reading and spelling.     

Dialect and authography: some differences between American and British spellers

Two experiments examined whether American and British university students make different kinds of spelling errors as a function of the differences between their dialects. The American students spoke a rhotic dialect, pronouncing an /r/ in such words as leper, hermit, horde, and gnarl. The British students, with their nonrhotic dialect, did not include an /r/ in such words. The dialect differences led to different spelling errors in the 2 groups. For example, the British students sometimes misspelled horde as "haud" because its vowel has the alternative spelling au in their dialect. They sometimes spelled polka as "polker" because its final vowel is often spelled as er in other words. The U.S. students were much less likely to make such errors, although they did make other errors that reflected aspects of their dialect. Phonology, far from being superseded by other strategies in the development of spelling, continues to be important for adults.     

Written spelling deficit of Broca''s aphasics

The purpose of this study was to investigate the nature of the written spelling deficit manifested by Broca's aphasics. Four spelling tests were given to eight Broca's aphasic patients. Analysis of misspelling errors led the investigators to conclude that Broca's aphasics do not spell phonically, but rather adhere exclusively to a visual/orthographic strategy. Possible cognitive deficits underlying the spelling problem are identified. The written spelling deficit is related to other features of the syndrome of Broca's aphasia, and finally, speculations are offered concerning the neurological substrate of written spelling in Broca's aphasic patients.     

Multiple levels of letter representation in written spelling: evidence from a single case of dysgraphia with multiple deficits

In this paper, we report a detailed analysis of the impaired performance of a dysgraphic individual, AD, who produced similar rates of letter-level errors in written spelling, oral spelling, and typing. We found that the distribution of various letter error types displayed a distinct pattern in written spelling on the one hand and in oral spelling and typing on the other. In particular, noncontextual letter substitution errors (i.e., errors in which the erroneous letter that replaces the target letter does not occur elsewhere within the word) were virtually absent in oral spelling and typing and mainly found in written spelling. In contrast, letter deletion errors and multiple-letter errors were typically found in oral spelling and very exceptional in written spelling. Only contextual letter substitution errors (i.e., errors in which the erroneous letter that replaces the target letter is identical to a letter occurring earlier or later in the word) were found in similar proportions in the three tasks. We argue that these contrasting patterns of letter error distribution result from damage to two distinct levels of letter representation and processing within the spelling system, namely, the amodal graphemic representation held in the graphemic buffer and the letter form representation computed by subsequent writing-specific processes. Then, we examined the relationship between error and target in the letter substitution errors produced in written and oral spelling and found evidence that distinct types of letter representation are processed at each of the hypothetized levels of damage: symbolic letter representation at the graphemic level and representation of the component graphic strokes at the letter form processing level.     

Writing nonsense: the interaction between lexical and sublexical knowledge in the priming of nonword spelling

The task of spelling nonwords to dictation necessarily requires the operation of a sublexical or assembled sound-to-spelling conversion process. We report an experiment that shows a clear lexical priming effect on nonword spelling (e.g., /vi:m/ was spelled as VEME more often following the prime word "theme" and as VEAM more often following "dream"), which was larger for lexically low-probability (or low-contingency) than for common (or high-contingency) spellings. Priming diminished when an unrelated word intervened between the prime word and target nonword and did so more for the production of low- than for high-contingency spellings. We interpret these results within an interactive model of spelling production that proposes feedback from the graphemic level to both the lexical and assembled spelling processes.     

Comparison between Oral and Written Spelling in Alzheimer's Disease

Written and oral spelling were compared in 33 patients with Alzheimer's disease (AD) and 25 control subjects. AD patients had poorer spelling results which were influenced by orthographic difficulty and word frequency, but not by grammatical word class. Lexical spelling was also more deteriorated than phonological spelling. Moreover, oral spelling was more impaired than written spelling in AD patients, whereas no difference was present between oral and written spelling of controls. Analysis of spelling errors showed that, for controls, errors were predominantly phonologically accurate in both spelling tasks. Significantly, AD patients produced more phonologically accurate than inaccurate errors in written spelling, whereas these errors did not differ in oral spelling. In contrast to controls who produced more constant than variable responses in oral and written spelling, AD patients made more variable responses (words correctly spelled in one task but incorrectly in the other) and they showed many instances of variable errors (different misspellings from one spelling task to the other). Two stepwise regression procedures showed that written misspellings were specifically correlated with language impairment, whereas oral spelling errors were correlated with attentional and language disorders. These results suggest that AD increases the attentional demands of oral spelling process as compared to written spelling. This dissociation argues, either for a unique Graphemic Buffer in which oral spelling requires more attentional resources than written spelling or for the hypothesis of separate buffers for oral and written spelling.     

Surface dyslexia

Two cases of surface dyslexia are described. In this disorder, irregular words such as broad or steak are less likely to be read aloud correctly than regularly-spelled words like breed or steam; and when irregular words are misread the incorrect response is often a regularisation (reading broad as "brode" and steak as "steek, for example). When reading comprehension was tested, homophones were often confused with each other: for example, soar was understood as an instrument for cutting, and route was understood as being part of a tree. Spelling was also impaired, with the majority of spelling errors being phonologically correct: for example, "search" was spelled surch. "Orthographic" errors in reading aloud (omitting, altering, adding or transposing letters) were also noted. These errors were not due to defects at elementary levels of visual processing.

One of our cases was a developmental dyslexic, and the other was an acquired dyslexic. The close similarity of their reading and spelling performance supports the view that surface dyslexia can cccur both as a developmental and as an acquired dyslexia.

A theoretical interpretation of surface dyslexia within the framework of the logogen model (including a grapheme-phoneme correspondence system for reading non-words) was offered: defects within the input logogen system, and in communication from that system to semantics, were postulated as responsible for most of the symptoms of surface dyslexia.     

The graphemic buffer and attentional mechanisms

Two patients with acquired dysgraphia were reported. The patients' performance in various written and oral spelling tasks converge in support of the hypothesis that they have selective damage, within the spelling system, to the Graphemic Buffer. Although the patients present with comparable patterns of error types, they differ in the distribution of errors as a function of letter position in words. The patients' patterns of errors are compared to previously reported patterns of spelling errors in dysgraphic patients and are discussed in terms of hypothesized mechanisms that operate on the representations that are stored in the Graphemic Buffer.     

Why simple verb forms can be so difficult to spell: the influence of homophone frequency and distance in Dutch.

Two experiments are reported in which the determinants of spelling errors on homophonous verb forms in Dutch were studied. Both experiments indicated that errors were determined by the frequency relationship between the two homophonous forms and the distance between the verb and the word determining its spelling. We propose an interference model for spelling in which a phonologically driven retrieval process is the locus of the frequency effect and a morphosyntactically driven computational component can account for the distance effect. Alternative explanations are also explored.     

Age-related influences on lexical selection and orthographic encoding during homophone spelling

Two experiments investigated age differences in how semantic, syntactic, and orthographic factors influence the production of homophone spelling errors in sentence contexts. Younger and older adults typed auditorily presented sentences containing homophone targets (e.g., blew) that were categorized as having a regular spelling (EW) or an irregular spelling (UE). In Experiment 1, homophones were preceded by an unrelated word, a semantic prime that was congruent with the target's meaning in the sentence (e.g., wind), or a semantic prime incongruent with the target's meaning (e.g., sky) and instead related to the competitor homophone. Experiment 2 manipulated the target's part of speech, where target and competitor homophones shared or differed in part of speech. For both age groups, significant semantic priming occurred, where homophone errors decreased following congruent semantic primes and increased following incongruent primes. However, priming only occurred when homophones shared part of speech. Further, both age groups made more errors on homophones with an irregular than a regular spelling, and this regularity effect was smaller for older adults when homophones shared part of speech. Contrary to many spoken production tasks, older adults made fewer errors overall than younger adults. These findings demonstrate age preservation in lexical selection but age differences in orthographic encoding, resulting in older adults producing fewer errors because of reduced activation to competitor homophones. These findings also illustrate that syntactic factors, such as part of speech, can influence the spellings of individual words.     

Oral reading in dementia

Reading has been thought to consist of three main processing components: the orthographic, phonological, and semantic lexicons. In traditional psycholinguistic models, these components have been treated independently such that the selective dysfunction of one does not necessarily imply the breakdown of another. Recently, it has been proposed that a word's semantic representation is essential to oral reading such that a disturbance within the semantic lexicon will disrupt processing within the orthographic and/or phonological lexicons. From this view, semantic deterioration should lead to fragmentation of the other systems contributing to reading, resulting in a specific pattern of errors during oral reading. This would include (1) a larger than normal advantage for reading words with regular spelling-to-sound correspondence over words with exception spelling, as well as the production of "regularization errors" when reading exception words; and (2) a smaller than normal difference between reading real words and pronounceable nonwords, or pseudowords (PW's). We found that patients with Semantic Dementia generally conformed to these hypothesized patterns of reading difficulty. Despite the presence of a semantic impairment, however, patients with Alzheimer's Disease, Frontotemporal Dementia, and Progressive Non-Fluent Aphasia did not demonstrate these patterns of reading difficulty. Our findings suggest that not all semantic impairments invariably lead to the disruption of the orthographic and phonological lexicons.     

Representation of letter position in spelling: Evidence from acquired dysgraphia

The graphemic representations that underlie spelling performance must encode not only the identities of the letters in a word, but also the positions of the letters. This study investigates how letter position information is represented. We present evidence from two dysgraphic individuals, CM and LSS, who perseverate letters when spelling: that is, letters from previous spelling responses intrude into subsequent responses. The perseverated letters appear more often than expected by chance in the same position in the previous and subsequent responses. We used these errors to address the question of how letter position is represented in spelling. In a series of analyses we determined how often the perseveration errors produced maintain position as defined by a number of alternative theories of letter position encoding proposed in the literature. The analyses provide strong evidence that the grapheme representations used in spelling encode letter position such that position is represented in a graded manner based on distance from both-edges of the word.     

An error analysis of misspellings made by learners of English as a first and as a second language

About 700 misspelled words were collected from the responses produced on a fill-in-the-blank task by two groups of advanced learners of English, (1) English-speaking children (9–11 years) and (2) Spanish-speaking adults studying English. The spelling errors were coded using a detailed categorization system (whose use and rationale are described), and the resulting tabulations were analyzed for differences between the two subject groups or differences across error types. The two groups were similar in that they both made proportionally more vowel than consonant errors. On the other hand, significant differences between the subject groups were found in three of the major categories: The Spanish speakers made more errors involving consonant doubling, while the native English speakers made more involving the unstressed vowel schwa (||) and the grapheme silent |e|. It is argued that these differences stem from the language backgrounds and resulting spelling strategies of the two groups, and the paper concludes with a discussion of the need for other studies comparing the spelling errors of first-and second-language learners.     

Effects of spelling errors on the perception of writers

The authors investigated the perceived relationship between spelling errors and cognitive abilities in a series of 3 experiments. Specifically, they examined whether college students' ratings of an author's intellectual ability, logical ability, and writing ability were affected by the presence of spelling errors. In the 1st experiment, the presence of 4 spelling errors in a short essay did not significantly affect the ratings. The spelling ability of college students, as measured by a standard oral dictation spelling test, was moderately conelated with a brief test of intelligence. In a 2nd experiment, college students rated the author of a short essay as having lower ability when there was a large number of spelling errors. The effect was more pronounced on the ratings of writing ability than it was on the ratings of logical ability or intellectual ability. This finding was replicated in a 3rd experiment, in which the essay contained misspellings actually made by writers. The results suggest that spelling errors can affect how people perceive writers, particularly when there are many spelling errors. College students appear to attribute spelling errors more to writing ability than they do to general cognitive abilities such as intelligence and logical ability.     

From letters to words: Procedures for word recognition in letter-by-letter reading

There are two alternative interpretations of how word recognition is accomplished in letter-by-letter reading. One postulates that it relies on the spelling system operating "in reverse," whereas the other claims that it is mediated by the reading system. Because of the close association between patterns of reading and spelling in previously reported cases of letter-by-letter reading, both hypotheses may be considered equally plausible. We now report a patient with letter-by-letter reading who demonstrated a striking dissociation between reading and spelling. Our observations in this patient argue against the concept of "reading via spelling" and suggest that word recognition in letter-by-letter reading is mediated by the orthographic input lexicon used in normal reading.     

Didn't You Run the Spell Checker? Effects of Type of Spelling Error and Use of a Spell Checker on Perceptions of the Author

We investigated expectations regarding a writer's responsibility to proofread text for spelling errors when using a word processor. Undergraduate students read an essay and completed a questionnaire regarding their perceptions of the author and the quality of the essay. We manipulated type of spelling error (no error, homophone error, non-homophone error) and information provided about the author's use of a spell checker (no information, author did not use a spell checker, author did use a spell checker). Participants' perceptions of the author's abilities and the quality of the essay suffered when the essay contained non-homophone spelling errors—errors that are typically flagged by a spell checker. Further, participants reported that they would be most likely to blame the writer rather than the spell checker for spelling errors contained in the text. These findings suggest that perceptions of both an author's abilities and the written products are affected by spelling errors. Even when supportive tools are available, the responsibility for producing error-free text remains with the author.     

Writing with the right hemisphere

We studied writing abilities in a strongly right-handed man following a massive stroke that resulted in virtually complete destruction of the language-dominant left hemisphere. Writing was characterized by sensitivity to lexical-semantic variables (i.e., word frequency, imageability, and part of speech), semantic errors in writing to dictation and written naming, total inability to use the nonlexical phonological spelling route, and agrammatism in spontaneous writing. The reliance on a lexical-semantic strategy in spelling, semantic errors, and impaired phonology and syntax were all highly consistent with the general characteristics of right hemisphere language, as revealed by studies of split-brain patients and adults with dominant hemispherectomy. In addition, this pattern of writing closely resembled the syndrome of deep agraphia. These observations provide strong support for the hypothesis that deep agraphia reflects right hemisphere writing.     

“Without the spelling errors I would have shortlisted her…”: The impact of spelling errors on recruiters’ choice during the personnel selection process

Despite the time spent on writing at work and employers’ dissatisfaction with their employees’ spelling skills, little is known about recruiters’ attribution, and decision making when they read application forms with spelling errors. This study examines the impact of spelling and typographic errors on recruiters’ attributions about applicants, and on their shortlisting decision. Based on a sample of 20 French recruiters, we conducted an experiment to collect both qualitative data through the verbal protocol method and quantitative data. Specific verbal reports are associated with different types of errors. Recruiters also form attributions about candidates. We demonstrate that spelling errors affect recruiters’ behavior more negatively than typographic ones. We discuss the implications of these findings for researchers and practitioners.     

Is comprehension necessary for error detection? A conflict-based account of monitoring in speech production

Despite the existence of speech errors, verbal communication is successful because speakers can detect (and correct) their errors. The standard theory of speech-error detection, the perceptual-loop account, posits that the comprehension system monitors production output for errors. Such a comprehension-based monitor, however, cannot explain the double dissociation between comprehension and error-detection ability observed in the aphasic patients. We propose a new theory of speech-error detection which is instead based on the production process itself. The theory borrows from studies of forced-choice-response tasks the notion that error detection is accomplished by monitoring response conflict via a frontal brain structure, such as the anterior cingulate cortex. We adapt this idea to the two-step model of word production, and test the model-derived predictions on a sample of aphasic patients. Our results show a strong correlation between patients’ error-detection ability and the model’s characterization of their production skills, and no significant correlation between error detection and comprehension measures, thus supporting a production-based monitor, generally, and the implemented conflict-based monitor in particular. The successful application of the conflict-based theory to error-detection in linguistic, as well as non-linguistic domains points to a domain-general monitoring system.